Getting to the Heart of Alzheimer’s Disease

Brain hypoperfusion in the pathogenesis of AD

It is commonly known that the brain relies on adequate blood supply (and thus oxygen) in order to maintain proper function²⁰. It has been proposed that hypoxia and ischemia resulting from cerebral hypoperfusion is a direct contributor to the pathogenesis and development of AD^20-22. Roher et al. observed total cerebral blood flow to be 20% lower in the AD group compared to the non-demented control group suggesting an association between brain hypoperfusion and dementia of AD²². The same study found a decrease in pulse pressure in AD patients, further indicating a decrease in cerebral blood flow, and consequently, a decrease in cognitive measures²². Nishimura et al. more specifically observed decreased cerebral blood flow to the frontal lobe in correlation to both reduced cognitive function and progression of AD²³. The reduced cerebral perfusion causes a metabolic energy crisis due to the reduced oxygen exposure. This hypoxic state induces oxidative stress and acidosis eventually culminating in neuronal degradation^{24, 25}.

Oxidative stress in AD

Contrary to popular belief, it has been discussed that a balanced reactive oxygen species (ROS) response actually promotes tissue repair via disinfection of existing tissue and stimulation of healthy tissue turnover²⁶. However, the imbalanced metabolism of oxygen induced by this hypoxic state, leads to excess ROS, resulting in deleterious effects²⁷. Furthermore, the mitochondria of the vascular wall cells have been identified as the primary target of oxidative stress prior to development and progression of AD²⁸.

However, it is believed that the accumulation of ROS, and thereby oxidative stress, on neural tissue is a result of hypoperfusion in combination with Aβ proteotoxicity²⁹. An increase in Aβ production has been linked to the progression of oxidative stress which can induce mitochondrial dysfunction. Work by Matsuoka et al.³⁰ using transgenic mice carrying mutant APP (amyloid precursor protein) and PSEN1 illustrates the key role of oxidative stress in the AD model. Additionally, 3-nitrotyrosine (protein oxidative stress marker), and 4-hydroxy-2-noneal (lipid oxidative stress marker) have been found to be increased with progressing levels of fibrillary Aβ³⁰. While most Aβ aggregations are found in extracellular regions, some collections of Aβ have been found in the mitochondria^{31, 32} of individuals with AD. This phenomenon may diminish mitochondrial function, specifically respiration and thereby increase the levels of ROS inside and potentially outside of the cell. Oxidative stress not only can induce further Aβ production, but triggers the production of tau protein, another critical component to the pathology of this disease³¹. Furthermore, it has been demonstrated that individuals with AD have decreased antioxidant levels- allowing higher levels of ROS to accumulate in the local environment²⁹. This evidence therefore suggests a cyclic effect between Aβ aggregation and oxidative stress, driving the progression of AD and its physical symptoms.

Inflammation contributes to vascular and neurodegeneration in AD

Another key aspect of the pathogenesis of AD as a result of reduced blood flow is inflammation. Changes in the vasculature either as a result of AD or part of an overall vasculopathy are associated with the release of multiple inflammatory factors, with cerebral micro vessels secreting higher levels of TNF-α, IL-1β, IL-6, and leukocyte adhesion molecules than non-AD controls³³. However, despite the vast amount of literature focused on neural inflammation and its role in AD, no clear evidence has illuminated whether inflammation is a contributing factor towards the etiology of AD, part of its pathology, or a secondary phenomenon³⁴. However, it is widely accepted that the negative consequences of inflammation contribute towards the progression of AD, including neurodegeneration and thereby diminished cognitive function. It is also understood that microglia, the resident macrophage in the brain, play a key role in the activation of the inflammatory response. Studies have illustrated that microglia are increased in individuals with AD and in transgenic mouse models Although they exhibit heterogeneous phenotypes, it is understood that they are involved in Aβ maintenance³⁵. Studies in vitro suggest that cytokine signaling and secretion from microglia can greatly impact the cerebral microenvironment and function of neurons. These cytokines - specifically IL-1β, IL-6, TNF-α, INF-γ - and various chemokines can strengthen the inflammatory response³⁵. More recently, cytokine signaling in AD mouse models can have significant effects on amyloidosis, neurodegeneration and cognition³⁵, potentially disrupting the blood brain barrier in neurodegenerative disorders³⁶. In AD pathology, the integrity of the blood brain barrier (BBB) is reduced^{37, 38} as well. Whether as a result of the chronic cerebral inflammation or part of a larger systemic pathology, this defective BBB is believed to play a role in the pathogenesis of AD.

Structural consequences of hypoperfusion in the AD brain

Furthermore, this same cerebral hypoperfusion has been found to breakdown the neurovascular unit (NVU), enabling progression into neurodegenerative disorders like AD³⁹. The NVU encompasses many different types of brain cells (endothelial, pericytes and vascular smooth muscle cells), which in turn control the blood brain barrier (BBB)³⁹. Breakdown of the NVU would result in a dysfunctional BBB. Many recent studies have now correlated BBB dysfunction with the accrual of vasculotoxic molecules and hypoxia resulting from a decrease in CBF³⁹.

Amyloid directly contributes to cerebral angiopathy

Changes in peripheral perfusion also have the potential to induce cerebral amyloid angiopathy (CAA) - fibrillary amyloid deposition in small cerebral vessels. However, the exact mechanism whereby CAA contributes towards the pathogenesis of AD remains unknown. It has been reported that CAA can affect cell viability, induce apoptosis or oxidative stress, and trigger an inflammatory response⁴⁰. Furthermore, CAA can create vascular dysfunction through hemorrhagic complications or blocking blood flow (resulting in cerebral ischemia). These events, either alone or in combination, may contribute towards the progression of AD⁴⁰.

Atherosclerosis, hypertension and ischemia cascade effects in AD

Additionally, for many decades, the role of decreased cerebral blood flow in the pathogenesis of Alzheimer’s relied on the impairment of vasculature through the lens of atherosclerosis^{14, 41}. Similarities between AD and atherosclerosis have sparked investigation considering that both are found in conjunction with vascular wall thickening and blood vessel occlusion⁴². After examination of the cerebral arteries in AD patients, Roher et al found cases of significantly increased cerebral artery occlusion in comparison to control groups^{42, 43}. Additionally, a positive correlation was determined between arterial stenosis and NFT⁴³, which is a hallmark of AD. Other studies by Roher et. al found more widespread intracranial atherosclerosis in AD patients versus nondemented patients ⁴⁴. Cognitive dysfunction was also found to be exacerbated in AD patients with cerebral atherosclerosis compared to non-atherosclerotic AD patients, regardless of intracranial or extracranial localization⁴⁵. Furthermore, the Nun Study of Aging and Alzheimer’s Disease indicated that patients with multiple brain infarctions have lower cognitive function and higher prevalence of dementia^{46, 47}.

Hypertension has been independently identified as a risk factor for AD, but many have reasoned that this is due to hypertension leading to other diseases, which then result in the onset of dementia⁴⁶. Throughout time, there have been multiple intuitions about the exact mechanism of hypertension leading to AD: (1) vascular alterations leading to infarcts in the brain, (2) adverse effects on neuronal health leading to deposition of Aβ in the brain, and lastly (3) development of cardiovascular disease giving rise to AD⁴⁶. Hypertension has also been proposed to evoke AD via ischemia, oxidative stress, inflammation and small-vessel disease^48,49. More recently, hypertension has been largely noted as the most detrimental vascular risk in the progression of AD⁵⁰. This designation is due to its causation of small-vessel disease, which later results in lacunar infarcts, white matter lesions/hyperintensities, and microinfarcts⁵¹; all of which are indicated in AD and accelerate the reduction in CBF⁵². Hypertension on the other hand is an independent risk factor for cardiovascular diseases including heart failure⁵³ indirectly affecting AD through the old lens of cerebral hypoperfusion and the new one of the common pathogenesis of AD and HF as age-related proteinopathies.

One of the main hallmarks of cerebrovascular disease is found in the microvascular structural changes, namely white matter hyperintensities⁵⁴, a manifestation of small-vessel disease⁵⁵. White matter hyperintensities (WMH) are a commonality with age progression and a causative agent behind normal cognitive decline⁵⁶. However, as imaging and other diagnostic techniques have progressed, it has been shown that small-vessel disease and WMHs in the brain both play a role in the development of AD⁵⁷. Population studies using MRI have also designated a high occurrence of small-vessel disease being associated with higher risk of stroke and dementia²⁰. These white matter hyperintensities are indicative of vascular lesions and have been found to lower the threshold for the clinical diagnosis of AD²⁰. Tosto et. al established that WMH distributions in the brain can progress to AD both directly via neurodegenerative changes and indirectly via aggravation of the tau effect on clinical transformation⁵⁶. This connection has enabled vascular risk factors to be common targets when testing novel therapeutic approaches against AD, such as hypertension and cholesterol treatments⁵⁸. However, these studies have been preliminary and there is much more to be investigated.

In summary, the role of the vasculature in Alzheimer’s Disease progression is one that is implicated primarily via decreased cerebral blood flow. This remains the main talking point for the vascular hypothesis of AD, which, as seen in Figure 1, claims that many vascular risk factors exert their pathology through cerebral hypoperfusion⁵⁹. However, there is a multitude of contributing vascular risk factors, some of which include atherosclerosis, hypertension, small-vessel disease, and BBB dysfunction, all contributing to reduced cerebral blood flow and the progression of AD, as shown in Figure 1. Much of the discussed literature has focused solely on the vasculature and the brain; however, it was recently proposed that there is a closer link between the brain and the heart than originally expected ⁶⁰.

Amylin cardiomyopathy: a direct link between heart and brain failure

Recent studies focusing on hyperamylinemia, a common disorder in diabetic patients, have revealed a more complex systemic pathogenesis of aggregating amylin. Previous work has found evidence of amylin/amyloid aggregations in the hearts of patients with diabetic cardiomyopathy⁷⁰. Work by Jackson et al.^{71, 72} sought to examine the cerebrovascular tissue of diabetic patients with vascular dementia of AD for amyloid deposits wherein they found amylin oligomers and plaques in the temporal grey matter from diabetic patients. Furthermore, researchers found amylin deposition in the brain vasculature and parenchyma of individuals in a specific test group with late onset AD and no apparent diabetes ⁷¹. Jackson et al.⁷¹ also found some instances where amylin and Aβ depositions were mixed. Expanding on this finding, it has been suggested that amylin and Aβ are connected in terms of a wider net pathophysiology. This hypothesis is supported by the fact that both Aβ and amylin can form similar functioning toxic aggregates that can induce inflammation, oxidative stress, and changes in the microvasculature of brain parenchyma^{70, 71, 73}. These findings suggest that amylin deposition and thereby its negative effects on the vasculature and parenchyma of the brain, may participate in the progression of AD.

Common genetic profiles between AD and HF

The concept of AD affecting both the head and heart is reinforced when looking at genetic profiles between connected disease states. Research has shown that there are similar genetic profiles between HF and AD. Work by Li et al.¹¹ found that in the familial forms, these two conditions share variations in the PSEN1 or PSEN2 genes. In this specific study, a PSEN1 missense mutation (Asp333Gly) and a PSEN2 missense mutation (Ser130Leu) were associated with both DCM and HF¹¹. Similarly, Gianni et al identified the same missense mutations in the PSEN1 and PSEN2 genes associated with AD in sporadic cases of iDCM and described new genetic variants of the promoter region of the genes affecting the expression levels of the protein¹². In this study Gianni et al. discovered plaque-like amyloid deposits in the hearts of patients with iDCM. These aggregations of proteins hold a considerable degree of proteotoxicity, inducing cell death ⁷⁴. At the micro scale, Demuro and colleagues investigated the biochemical mechanisms by which aggregations of soluble amyloid proteins have a neurotoxic effect⁷⁵. Their findings suggest that oligomer amyloid aggregations can disrupt calcium flux homeostasis and thereby cell membrane function⁷⁵. Similar to this proteotoxic effect in neurons, oligomeric aggregates exercise the same effect on cardiomyocyte Ca²⁺ homeostasis¹². The toxic role of amyloid deposition as a causal agent for AD was challenged by the failure of some clinical trials targeting amyloid plaques for efficacy of clinical symptoms (the immunoglobulin/albumin combination Flebogamma/Albutein, and small molecule targeting the beta-secretase 1 cleaving enzyme Verubecestat) or causing side effects such as encephalopathy (the humanized anti-Aß mAb Ganatenerumab). An initial explanation for the failure of the trials, and therefore the failure of the “amyloid theory” altogether, included late administration of the drug when the amyloid had triggered neuronal cell death and other terminal changes. However, more recent trials targeting multiple forms of Aß, such as oligomers in addition to the insoluble fibrils (the mAb Crenezumab, Aducanumab) did not cause side effects and provided some evidence of clearance of plaques and decline of clinical symptoms. While waiting for the ongoing phase III clinical trial results (e.g for Crenezumab, Aducanumab, the BACE inhibitors Lanabecestatand Elenbecestat) in vitro studies from the del Monte laboratory using atomic force microscopy suggested that a possible explanation for the failure of some of the drugs may reside in targeting fibers vs. the more toxic oligomeric species. The study indicated that while dissolving fibers may result in increased release of toxic species, targeting the latter may at least slow the progression of the disease by removing the source of neuronal poison⁷⁶.

Additional common protein profiles between AD and HF

Focusing on the heart, the del Monte laboratory analyzed the composition of the aggregates in the myocardium of iDCM patients and identified that cofilin-2 - along with its substrate actin and competing protein MLCII - was sequestered in the aggregates⁷⁷. Cofilin is an actin-depolymerization protein that regulates the turnover of actin in contractile cells and the structural integrity of the cell. In a defective and sequestered state, this protein is also known to play a role in multiple neurodegenerative diseases such as corticobasal degeneration, William’s syndrome, fragile X syndrome, and spinal muscular atrophy^{12, 60, 77} as well as other cardiac disorders such as myocardial ischemia. Recently, the Salloum laboratory described similar changes in cofilin activity, as described in iDCM patients with end stage ischemic cardiomyopathy⁷⁸. This new evidence further supports the link between neurodegenerative disease and various cardiovascular diseases leading to HF.

Following this work, Dr. del Monte’s group analyzed the complex pathology of proteinopathies and sought to discover a closer link between HF and AD by examining the hearts of individuals with AD diagnosis. In this work, Troncone et al. analyzed the hearts and brains of patients with AD. Upon investigation, they discovered, in the heart, Aβ (both Aβ₄₀ and Aβ₄₂) structurally akin to those found in the brain⁶⁰ and, in a retrospective analysis, found that AD patients present with myocardial diastolic dysfunction. Thus, like in traditional cardiac amyloidosis, the pathophysiology of diastolic dysfunction in AD can be described by the accumulation of misfolded proteins in the heart ⁷⁹.

The concept of peripheral accumulation of Aβ in patients with AD is not restricted to the heart. In fact, an early study by Joachim, Mori, & Selkoe⁸⁰ analyzed Aβ deposition in non-neuronal tissue - most notably, skin and intestine. Eight of the samples from test subjects with AD showed clear and definite evidence of Aβ deposition in these peripheral tissues. Similarly, Aβ₄₀ and Aβ₄₂ aggregates were identified in skeletal muscle of AD individuals using fast performance liquid chromatographic (FPLC) size exclusion chromatography. Researchers discovered elevated levels of Aβ in the temporalis muscles of AD individuals, indicating a possible contributor to elevated concentrations of Aβ plasma levels and potentially indirectly contributing to Aβ deposits in cerebral blood vessels and brain parenchyma⁸¹. Furthermore, these varying levels of Aβ suggest an alteration in APP and/or Aβ metabolism in peripheral tissues outside of the CNS⁸¹.

Viewing Alzheimer’s disease with a new lens as a potential multi-organ disease⁶⁰, as shown in Figure 2, it is possible that an inflamed and acidic cerebral microenvironment potentially contributes to a defective BBB (a common symptom of AD). This impaired BBB allows permeability of Aβ plaques into the bloodstream. This spreading event in combination with a defective production/clearance homeostasis of misfolded proteins would cause a combination of central nervous system (CNS) and peripheral symptoms. In the case of HF, the involvement of the heart would accelerate the progression of the disease by contributing to the oxidative stress and acidosis via brain hypoperfusion. On the other hand, there might be a possible specific genetic profile- including a miRNA signature - that accounts for the systemic link between the head and the peripheral organs.